Research map / 3D vision / Face and shape
VertexRegen: Mesh Generation with Continuous Level of Detail
Xiang Zhang, Yawar Siddiqui, Armen Avetisyan, Chris Xie, Jakob Engel, Henry Howard-Jenkins
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add progressive-mesh (vertex split) tokenization versus EdgeBreaker/flattened tokenization | VertexRegen · unconditional generation | JSD: 3.22 -> 2.89 | Improved |
| compare VertexRegen with MeshXL on COV and MMD | VertexRegen · unconditional generation | COV / MMD: 51.76 -> 51.03 | No clear effect |
| generate with continuous level of detail instead of face-count condition token (MeshXL w/ FCC) | VertexRegen · 400 face constraint | 1-NNA: 59.06 -> 51.03 | Improved |
| same comparison at 400 face constraint | VertexRegen · 400 face constraint | COV: 41.20 -> 50.92 | Improved |
| add half-edge data structure to vertex split tokenization (ablation) | VertexRegen · tokenization compression | compression ratio: 0.89 -> 0.73 | Improved |
| add guided (geometry-constrained) decoding (ablation) | VertexRegen · unconditional generation | JSD: 3.37 -> 2.89 | Improved |
| add guided decoding (ablation) | VertexRegen · unconditional generation | average faces |F|: 211 -> 320 | Improved |
| add guided decoding (ablation) | VertexRegen · unconditional generation | COV: 51.12 -> 51.03 | No clear effect |
| add guided decoding (ablation) | VertexRegen · unconditional generation | 1-NNA: 50.75 -> 50.22 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MeshAnything V2 — MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 2025) · better on 3 of 4 shared results
- MeshXL — MeshXL: Neural Coordinate Field for Generative 3D Foundation Models (NeurIPS 2024) · better on 2 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ARMesh: Autoregressive Mesh Generation via Next-Level-of-Detail Prediction (NeurIPS 2025 · similarity 0.90)
- TreeMeshGPT: Artistic Mesh Generation with Autoregressive Tree Sequencing (CVPR 2025 · similarity 0.89)
- MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 2025 · similarity 0.88)
- MeshGen: Generating PBR Textured Mesh with Render-Enhanced Auto-Encoder and Generative Data Augmentation (CVPR 2025 · similarity 0.87)
- Scaling Mesh Generation via Compressive Tokenization (CVPR 2025 · similarity 0.87)
- CraftsMan3D: High-fidelity Mesh Generation with 3D Native Diffusion and Interactive Geometry Refiner (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
